• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.39 no.2
• /
• pp.39-49
• /
• 2002

This paper proposes the MAWUPC (Motion Adaptive Weighted Unmatched Pixel Count) algorithm to track multiple objects of similar color The MAWUPC algorithm has the new method that combines color and motion effectively. We apply the MAWUPC algorithm to face and hand tracking against complex background in an image sequence captured by using single camera. The MAWUPC algorithm is an improvement of previously proposed AWUPC (Adaptive weighted Unmatched Pixel Count) algorithm based on the concept of the Moving Color that combines effectively color and motion information. The proposed algorithm incorporates a color transform for enhancing a specific color, the UPC(Unmatched Pixel Count) operation for detecting motion, and the discrete Kalman filter for reflecting motion. The proposed algorithm has advantages in reducing the bad effect of occlusion among target objects and, at the same time, in rejecting static background objects that have a similar color to tracking objects's color. This paper shows the efficiency of the proposed MAWUPC algorithm by face and hands tracking experiments for several image sequences that have complex backgrounds, face-hand occlusion, and hands crossing.

• Proceedings of the IEEK Conference
• /
• 2001.09a
• /
• pp.341-344
• /
• 2001

본 논문은 두 대의 카메라로 제안하는 물체의 색상, 움직임, 형태상의 특성을 이용하여 3차원 공간상의 움직임을 실시간으로 추출하는 것을 목적으로 한다. 본 논문에서 제안하는 물체는 구조상 물체 자체가 캘리브레이션 물체의 역할을 포함하여 캘리브레이션이 되지 않은 상황에서도 정확하게 물체의 3차원 정보를 추출할 수 있으므로 3차원 입력 디바이스로 이용할 수 있다. 3차원 움직임을 추출하기 위해 먼저 3차원 공간상의 물체와 좌우 영상의 상관관계를 구하고 좌우 즉 영상에서 원이 위치할 탐색영역은 MAWUPC 알고리즘을 이용하여 예측한다. 탐색영역 안에서 PCA를 사용하여 원의 정확한 위치를 찾으며 좌우 영상에서 얻은 원의 위치와 스테레오 카메라의 기하학적 구조를 종합하여 3차원 움직임을 추출한다. 추출한 3차원 움직임은 가상환경에서 가상 물체의 움직임을 제어하는데 응용할 수 있다.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.39 no.4
• /
• pp.429-441
• /
• 2002

This paper concerns extracting 3D motion information from a 3D input device in real time focused to enabling effective human-computer interaction. In particular, we develop a novel algorithm for extracting 6 degrees-of-freedom motion information from a 3D input device by employing an epipolar geometry of stereo camera, color, motion, and structure information, free from requiring the aid of camera calibration object. To extract 3D motion, we first determine the epipolar geometry of stereo camera by computing the perspective projection matrix and perspective distortion matrix. We then incorporate the proposed Motion Adaptive Weighted Unmatched Pixel Count algorithm performing color transformation, unmatched pixel counting, discrete Kalman filtering, and principal component analysis. The extracted 3D motion information can be applied to controlling virtual objects or aiding the navigation device that controls the viewpoint of a user in virtual reality setting. Since the stereo vision-based 3D input device is wireless, it provides users with a means for more natural and efficient interface, thus effectively realizing a feeling of immersion.